Modelling of compacted bentonite swelling accounting for salinity effects

Abstract

This paper presents a formulation to incorporate the influence of water salinity on the swelling behaviour of a MX-80 bentonite into previously developed hydro-mechanical models that can reproduce swelling under dilute conditions. The effects of salinity on macro-and microstructural water chemical potentials were introduced. In addition, a description of solute transport was included to characterise the evolution of the system's salinity. A simplified geochemical model was adopted to idealise the geochemical complexity of bentonite. In addition, the modelling of the destructuration process that occurs during swelling was modified to account for the effect of salinity. The formulation was implemented in a multiphysics partial differential equation finite element solver, and the numerical model was used to simulate several vertical free swelling tests with feed water of different salt contents (deionised, 10, 35 and 70 g/L). The results demonstrate that even though the model can be developed further, it represents a significant improvement over models that do not account for the effects of salinity.

Comparison of the hydrogeochemical and mechanical behaviours of compacted bentonite using different conceptual approaches

“…The first of the numerical models considered in this study is the CRUNCHFLOW model (Steefel et al, 2015), which is a geochemical transport model with a simplified auxiliary geomechanical model implemented in the CRUNCHFLOWMC version; this model is referred to as the mC model in this study. The second model, developed in a Comsol Multiphysics environment, was proposed by Navarro et al (2017), which is primarily mechanical with a simplified geochemical model; this model is referred to as the Mc model in this study. Both models use a double-porosity approach to describe the compacted bentonite system.…”

“…First, the model Mc includes a formulation of the water exchange kinetics between micro-and macroporosities that considers the chemical potential difference between both porosity levels (Navarro et al, 2016). The model also uses a constitutive model of the behaviour of the microstructure that is based on the interpretation of retention curves at high suction values (Navarro et al, 2015) and incorporates an alternative definition of the interaction function between the micro-and macrostructural strains to consider the free swelling phenomena and the effect of the salinity of the saturating water (Navarro et al, 2017).…”

“…The description of the mechanical approach of the Mc presented in this work is based on the formulations of Hoffmann et al (2007) and Alonso et al (2011) for the behaviour of the macrostructural level. The conceptual approach for the behaviour of the microstructure and its influence on the macrostructure, has been taken from different works on expansive clays in free swelling conditions (Navarro et al, 2014;Navarro et al, 2015a;Navarro et al, 2016) and that include salinity effects on the microstructural behaviour (Navarro et al, 2017). This appendix only summarises the formulation Fitting coefficient for the mass transfer interaction Navarro (2014) emR Additional microstructural void ratio Navarro (2015)…”

Development of a THMC code for bentonites in COMSOL Multiphysics

“…In this context, the main objective of this work is the development of a THMC code that allows predicting the behaviour of the bentonite barrier from an unsaturated state during an extensive timeframe. For this purpose, the model of Navarro et al (2017) [13] has been improved by the implementation of an extended geochemical model typical of MX-80 bentonite [14] and the surrounding groundwaters. The code has been implemented in COMSOL Multiphysics.…”

Scale effect on hydraulic conductivity and solute transport: Small and large-scale laboratory experiments and field experiments

“…Numerical models are tools used in the geotechnical and geoenvironmental practice to solve a wide range of problems related to water flow and solute movement in the subsurface (Dou et al, 2014;Ghiglieri et al, 2016;Navarro et al, 2017;Wang et al, 2017). Hydraulic conductivity (K) and the solute transport parameters such as hydrodynamic dispersion coefficient (D), dispersivity (α), partition coefficient (Kd), and retardation factor (R) are key input parameters for these numerical models and their proper determination is fundamental (Bouchelaghem and Jozja, 2009).…”